## 
## Attaching package: 'dplyr'
## The following objects are masked from 'package:stats':
## 
##     filter, lag
## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union
## 
## Attaching package: 'plotly'
## The following object is masked from 'package:ggplot2':
## 
##     last_plot
## The following object is masked from 'package:stats':
## 
##     filter
## The following object is masked from 'package:graphics':
## 
##     layout
## Warning: NAs introduced by coercion

Introduction

Earthquakes are a force to be reckoned with. They’re powerful jolts caused by movement underneath the surface that can drastically change our world and environment. For this project, I will be analyzing a database that records significant earthquakes (5.5+ magnitude) from 1965 to 2016. The entire dataset is compiled by The National Earthquake Information Center (NEIC). Which takes data from national and international agencies, scientists, critical facilities, and the general public and complies the information gathered. The main goal of this dataset is to provide another resource for scientists researching trends and networks of global seismographic reports. But also utilized for public knowledge and interpretations since the source is on kaggle and accessible to everyone. Here is the link to the database

Data + Discussion

Looking at the various plots of the graph can display a lot of information that can often be looked over or ignored when looking just at the dataset, since it is very dense. Using plots can summarize the data collected and make the data visualizable. Since the information was very dense, I have made a couple plots to understand most aspects of the dataset.

The first visualization is to answer the question: Are the number of earthquakes increasing every year? To answer, I added the number of earthquakes every year and plotted the values on a line graph. This enables readers to see the trend of number of earthquakes each year. From first glance, it looks like the number of earthquakes fluctuates randomly where one year is 100 more or less than the previous year. However, looking carefully, there is an upward trend. Which might mean that future years might have a large amount of earthquakes each year. And this could affect more than the people living on the fault line. However, we must also ask the question of whether this information is reliable. The technology in the 1960’s might not have detected every significant earthquake due to the lack of sophisticated technology, so it is up to us to decide if it’s accurate. Nevertheless, there is always around 500 earthquakes each year so citizens be warned!

caption: displays location of earthquakes

This plot displays the locations of earthquakes, which is important when trying to understand where to be more cautious and predict future locations of earthquakes. Back then, researchers were able to determine the location of fault lines by these earthquake recordings. So highlighting these locations might uncover a new fault line or discovery. It also informs citizens in the areas to be prepared for earthquakes since there has been a history of earthquakes.

Reflection

Overall, this extra analysis has pushed me to learn more about R and about a topic that has impacted my life. Growing up in California, we regularly had earthquakes, both small and large. Understanding the reasons why and when can be really beneficial since as a child, I thought the world was going to crumble when an earthquake hit. It was a really neat experience to relate data science to something I care about and learn more about the topic. I also feel like I learned a lot more about R and practiced the skills I learned in the course on my own. There is much more that meets the surface, both technically and information based. There is always more to the information presented, like we saw here with the dataset and visualizations. And technically, especially when coding in R. We only hit the surface but there is a lot more that we haven’t explored. And doing this project has encouraged me to take on my own projects, since I was able to do this project without much help.

My code for the data analysis

df <- read.csv(“database.csv”) ### Year breakdown year <- df %>% mutate(year = as.numeric(substring(df$Date, 7,11))) %>% group_by(year) %>% count() officalYear <- year[complete.cases(year), ]

year_breakdown <- ggplot(officalYear, aes(x=year, y=n)) + scale_x_continuous(breaks=seq(1965, 2016, 5)) + geom_line(group = 1) + labs(x = “year”, y = “# of earthquakes”, title = “Number of Earthquakes every year”, caption = “(based on data from earthquake database. which only counts significant earthquakes that have a magnitude of 5.5 or higher.)”)

Map of earthquake locations

map <- list( scope = “world”, showland = TRUE, landcolor = toRGB(“gray95”), countrycolor = toRGB(“black”) )

map <- plot_geo(df, lat = ~Latitude, lon = ~Longitude) %>% add_markers( text = ~paste(“State:”, Magnitude), colors = c(“yellow”, “orange”, “red”, “red2”, “red3”), color = ~Magnitude, symbol = I(“circle”), hoverinfo = “text”, marker = list(size = ~Magnitude)) %>% colorbar(title = “Magnitude”) %>% layout(title = “Locations of Earthq”, geo = map)